Development of a quantitative semi‐mechanistic model of Alzheimer's disease based on the amyloid/tau/neurodegeneration framework (the Q‐ATN model)

Abstract Introduction A quantitative model of Alzheimer's disease (AD) based on the amyloid/tau/neurodegeneration biomarker framework (Q‐ATN model) was developed to sequentially link amyloid positron emission tomography (PET), tau PET, medial temporal cortical thickness, and clinical outcome (Clinical Dementia Rating – Sum of Boxes; CDR‐SB). Methods Published data and biologically plausible mechanisms were used to construct, calibrate, and validate the model. Clinical trial simulations were performed for different anti‐amyloid antibodies, including a 5‐year simulation of subcutaneous gantenerumab treatment. Results The simulated time‐course of biomarkers and CDR‐SB was consistent with natural history studies and described the effects of several anti‐amyloid antibodies observed in trials with positive and negative (or non‐significant) outcomes. The 5‐year simulation predicts that the beneficial effects of continued anti‐amyloid treatment should increase markedly over time. Discussion The Q‐ATN model offers a novel approach for linking amyloid PET to CDR‐SB, and provides theoretical support for the potential clinical benefit of anti‐amyloid therapy. Highlights A semi‐mechanistic model was developed to link amyloid/tau/neurodegeneration biomarkers to clinical outcome (Q‐ATN model). The Q‐ATN model describes the disease progression seen in natural history studies. Model simulations agree well with mean data from the aducanumab EMERGE study. A 5‐year simulation of gantenerumab predicts greater benefit with longer treatment.